Changing the ray version is not allowed

How severe does this issue affect your experience of using Ray?

  • High: It blocks me to complete my task.

I am trying to run a job on a local cluster

my runtime_env contains:

pip:
 - autogluon==0.5.2

I’m currently testing with just a head node and no workers (for debugging purposes).

I have Ray 2.0.1 on the head node.

In my config.yaml, I have this image: rayproject/ray:2.0.1-py39-cpu

If I open a bash to my docker container and check the ray version, it is 2.0.1

Where could the ray version 1.13.0 be coming from?